Question

I’m trying to pull in an element from an external website using PHP and cURL.

• The link to the website I’m trying to pull content from is:
  http://www.stayclassy.org/fundraise?fcid=231864

• The element I’m targeting is the number value under the list item
  “Raised So Far” in the right column at the top (right now the value is at $10).

• Here is the code I’m using to extract the data:

      define("TARGET", "http://www.stayclassy.org/fundraise?fcid=231864");
  
  
  $curl = curl_init();
  curl_setopt($curl, CURLOPT_URL, TARGET);
  curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
  
  if(!($results = curl_exec($curl))) {
      print("{ \"total\": \"$0.00\" }");
      return;
  } 
  
  $pattern = '/\<li class="goalTitle"\> \$(.+?) \<\/li\>\<\/a\>/';
  preg_match_all($pattern, $results, $matches);
  
  $total = $matches[1][0];
  $total = str_replace(",", "", $total);
  
  printf("{ \"total\": \"$%s\" }", formatMoney($total, true));
  
  
  function formatMoney($number, $fractional=false) 
  {
      if ($fractional) {
          $number = sprintf('%.2f', $number);
      }
      while (true) {
          $replaced = preg_replace('/(-?\d+)(\d\d\d)/', '$1,$2', $number);
          if ($replaced != $number) {
              $number = $replaced;
          } else {
              break;
          }
      }
      return $number;
  } 
  

The issue I’m having is that the list item/element I’m targeting doesn’t have a unique ID or class. In fact, the dollar amount is located in a separate list item without a class.

I was wondering how to target a specific list item in an unordered list using the code above, particularly when it doesn’t have a class. Any ideas?

Answer 1

Targeting the specific item requires that you identify a unique string around it. To do this you just expand further and further out until you find a string you can identify that only occurs once. So, the line you want is:

<li>$10</li>

but this is not unique at all. So we expand the string by adding the previous line as well:

<li class="goalTitle">Raised so far:</li>
<li>$10</li>

and bingo, this string is unique for your needs. The string is fairly constant except for your amount, so it will be easy to use. So you need a regular expression that finds this string. I’d use something like this:

$pattern = '/<li class="goalTitle">Raised so far:<\/li>\s*<li>\$(\d+)<\/li>/';

You don’t need to use preg_match_all because you only expect to get one match:

preg_match($pattern, $results, $matches);
$total = $matches[1];

---

Your other options include loading the page with a DOMDocument, and then using XPath or getElementById to parse the DOM. But that may be a little too much effort for this task.

Also, I’d use file_get_contents to fetch the contents of the remote site. But that’s just me.

---

UPDATE: To handle thousands separators as well, modify your pattern as follows:

$pattern = '/<li class="goalTitle">Raised so far:<\/li>\s*<li>\$([\d\.,]+)<\/li>/';